Energy Architecture methodology

When a client engages us to provide our Energy Architecture consulting service, our strategic thinking is set out in an Energy Architecture Paper. This is highly tailored to the client and it sets out our recommendation as a high-level design. 

For us, an important part of what we deliver for our clients is to go beyond simply offering some options to a client. Based on our judgement, the data, lots of discussion with our client, and lots of debate within the team, we propose an answer that we are comfortable defending.

However, we recognise that is not the answer. Partly that’s because each client’s plans and views evolve, often iteratively with our work. But it’s also a reflection of the complexity and data requirements for modelling an energy system that has not yet been built.

This can be an iterative, complex process, but it does not need to be a black box. For those who want to look inside, here’s how it works.

Finding system concepts

For us, a ‘system concept’ is a high-level description of a credible design option for an energy system. Within that concept, there will be many things that are, at this stage, unknown, undefined or implicitly assumed.

For any given situation, there are many possible system concepts. That’s not just a function of the range of technology options; it’s also because energy now has to be viewed as a ‘whole-system’, taking account of economics, operations, data and people. Just defining and describing the options could be a frustrating and circular task.

So we use morphological analysis, a problem-solving approach developed by Fritz Zwicky, the discoverer of dark matter. But because we’re not astrophysicists, we use a rather simplified version. This involves setting out some key parameters involved with the problem, and some options for each parameter. For instance, how electrical power is provided might be one parameter; with three options being grid-only, on-site generation only (an island microgrid) or a hybrid of the two.

The parameters and options used are specific to each client and situation, but to illustrate, let’s assume that there are three parameters, each with three options. This results in 27 possible system concepts, which can be thought of like a Rubik’s cube:

Each small cube represents a combination of settings for each of the three parameters. Cube A, for example, is a system concept that has electricity provided by the grid, powering all transport needs and with a hybrid heating system (perhaps heat pumps with a biomass back-up).

In this case, none of the combinations are impossible (the first way to rule out options). However, if the client specifies a strong preference against gas, then we can safely rule out the left-hand options (shown in grey below). Further, if the client’s site is not suitable for a separate microgrid, then we can rule out the top layer (the six cubes in blue that have not already been ruled out). With an understanding that at least some electric vehicles are desired, we can rule out the option of transport being only petrol-based and a further four options (in red) can be removed. Suddenly our 27 options are now down to a more manageable eight. Practically, let’s assume that we can also rule out a purely electric heating and transport system that is grid-powered, based on reliability and likely economics (the orange cube).

Selecting a system concept

Having simplified the problem into seven system concepts, we then assess these concepts against each other. We do this initially with a ‘Pugh matrix’. This starts by setting out a handful of criteria that are important to the client and applying a weight to each of them.

Finally, each system concept is judged as worse, roughly the same, or better (scored as -1, -0.5, 0, 0.5 or 1) than a defined reference case (usually the status quo). In this case let’s assume it is grid power, grid gas and petrol cars, i.e. the bottom back left cube. That judgement is partly based on experience, and partly based on our initial techno-economic modelling. Those scores, multiplied by the weights, gives an overall score for each system concept.

In this instance, the initial favourite is a fully electrified set-up.

Iterative refinement

From that starting point, we then consider the impacts from adding or removing a given element. We do this by mapping out the system with a large sheet of paper and a set of cards. Each card represents an energy asset, like an on-site wind turbine used to generate electricity.

It may sound simple, but it’s the best way that we’ve found to avoid circular debates or to get lost in spreadsheets. By running through each element in turn, starting from the higher-level decisions and moving to the smaller, we fill out our overall set of recommendations. For example, centralised heat generation with a heat network, vs. distributed heat generation, is a mid-level architectural decision that impacts overall system effectiveness and would determine the parameters for various lower-level choices.

For some companies, the decisions are intrinsically tied up with their overall business, so it makes sense to include them in some of these exercises. In other cases, we can update them afterwards. In either case, it is crucial for us that our clients feel that the process is unbiased and that they can have confidence in the results.

Stress testing

Once we have a detailed system plan, we try and break it. We model a wide range of risks and their potential impact. For example, how would the system operate with a given number of days of ‘dunkelflaute’ (when both wind and solar generation are ineffective)? How does it perform across the seasons? What’s the risk of a hotel guest having to take a cold shower?

Our basic principle here is that our clients’ first concern is making sure things work properly. Everything else comes after.

Modelling the outcomes

If our system has made it through this far, it’s time for the final tests. What’s the impact on decarbonisation? And what’s the impact on finance and company valuation? We test against what the client is looking for. That could mean minimising running costs whilst satisfying regulations. Or it could mean maximising valuation through improved profitability and carbon reductions.

Typically we measure carbon changes against the reference case we defined earlier. Financial and valuation impacts are compared to our client’s current forecasts (ideally on the same basis as our reference case). Our team contains financial and valuation experts, but we recognise our role here is to support our clients, not to replicate their work. We’re therefore focused on the change that we can bring, whether that’s carbon or cash saved, net present value added, or IRR % increased.

Presenting the result

Our detailed concept is then set out in our Energy Architecture Paper. This is a long-form report that brings the operational, commercial and investment aspects together in one place. It allows informed decision-making around a clear plan, and leads naturally on to the next stage, implementation.